Difference in measurements between left and right speakers (newbie)

[behindthesofa]

Newbie here, so please forgive me if I aslk stupid questions ;-)

So I recently got a UMIK-1 and am measuring my speakers to perform room correction. I did some periodic pink noise sweeps (moving microphone method) of the left and right channel separately. I think I see quite substantial differences in frequency response due to room modes between the two speakers, but that is not unexpected as they are not symmetrically placed in the room (left speaker much closer to the sidewall than right speaker).

This is not the issue I want to talk about though. I noticed a substantial dip @ 2.3kHz in the right speaker only. Can this be due to differences in crossover? I would be interested in any input on these differences. Should I convince my wife it's time for new speakers ?

Left measurement (avg of 3 mmm measurements) in orange, right measurement (avg of 4 mmm measurements) in blue/purple. I added the REW .mdat file as well with all the measurements.

Thanks a lot!

 

[Weeb Labs]

It is perfectly normal for each channel to produce a slightly different measurement depending upon placement.

Your right channel does appear to exhibit a very narrow cancellation at 2.3KHz but this is most likely the result of differing reflections and absorption; especially if the speaker is in close proximity to a wall or floor. It is unlikely to be audible and I would recommend using 1/12th or 1/24th smoothing.

Of much greater concern to me are the +15dB modal peaks around 70Hz. Those are definitely audible and should be pulled down with a few PEQ filters.

I would correct the full response but this is not everybody’s preference and results are dependent upon the smoothness of your speakers’ directivity.

 

[behindthesofa]

Thanks very much! I already created a PEQ filter to dampen those modal peaks, they indeed are certainly audible and I specifically target them with DSP. UMIK-1 + CamillaDSP are really cool. I will ignore the 2.3kHz dip then.

 

[ozzy9832001]

You'll get differences between the channels normally, unless everything is completely symmetrical, which it rarely is. I'd try PEQ filters with 1/6, 1/12 and 1/24 smoothing for the bass and see which ones you like the best. My guess is 1/6 will be fine. Below about 80hz will be hard to localize, so you can just use a filter for both channels. Above you can start to get localization especially with modal issues, so you may want to do channel by channel.

The 2.3khz dip could be desk or chair if they were present during the measurement. Slight change in position could help with that. Could also be small differences in speaker distance from the mic.

I wouldn't correct for it as it's likely not audible.

 

[behindthesofa]

Thanks for the detailed tips @ozzy9832001!

 

[droid2000]

It is perfectly normal for each channel to produce a slightly different measurement depending upon placement.

Your right channel does appear to exhibit a very narrow cancellation at 2.3KHz but this is most likely the result of differing reflections and absorption; especially if the speaker is in close proximity to a wall or floor. It is unlikely to be audible and I would recommend using 1/12th or 1/24th smoothing.

Of much greater concern to me are the +15dB modal peaks around 70Hz. Those are definitely audible and should be pulled down with a few PEQ filters.

I would correct the full response but this is not everybody’s preference and results are dependent upon the smoothness of your speakers’ directivity.

Quick question - I have no reference to determine what is a "narrow" or "wide" gap or peak. In this case you consider an almost 1K Hz gap around 2K Hz to be narrow? Is there a heuristic I can use to evaluate the audibility of gaps and peaks?

 

[keks8430]

from https://www.harman.com/documents/AudioScience_0.pdf

So, it depends on the signal type and the quality factor of the resonance.
You should (almost) never boost narrow gaps, for several reasons.